{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:5UQAZA3UREC53YR5GNP4UQCAWV","short_pith_number":"pith:5UQAZA3U","schema_version":"1.0","canonical_sha256":"ed200c83748905dde23d335fca4040b5576644f957f0b8165fd5464f9d2e469b","source":{"kind":"arxiv","id":"2210.04311","version":1},"attestation_state":"computed","paper":{"title":"Pruning Adversarially Robust Neural Networks without Adversarial Examples","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR","cs.CV"],"primary_cat":"cs.LG","authors_text":"Jennifer Dy, Stratis Ioannidis, Tong Jian, Yanzhi Wang, Zifeng Wang","submitted_at":"2022-10-09T17:48:50Z","abstract_excerpt":"Adversarial pruning compresses models while preserving robustness. Current methods require access to adversarial examples during pruning. This significantly hampers training efficiency. Moreover, as new adversarial attacks and training methods develop at a rapid rate, adversarial pruning methods need to be modified accordingly to keep up. In this work, we propose a novel framework to prune a previously trained robust neural network while maintaining adversarial robustness, without further generating adversarial examples. We leverage concurrent self-distillation and pruning to preserve knowledg"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2210.04311","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-10-09T17:48:50Z","cross_cats_sorted":["cs.AI","cs.CR","cs.CV"],"title_canon_sha256":"1831887aeb0c9bcac66a7febb9def84a735e0f65cc6f917bb92c78851768ae4e","abstract_canon_sha256":"6628cf44528125a28175efc74772e6cc1322a38de439ff477f1b6a80b44ff617"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:04:51.979268Z","signature_b64":"4fw8aiMD7l4ZiSicV27hOmFtoYBUyvvPxoS9Edqvt0kxP40ax/QnxsEfzsUdYWHxrXkoq0i/LijlCJL14NdMDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed200c83748905dde23d335fca4040b5576644f957f0b8165fd5464f9d2e469b","last_reissued_at":"2026-07-05T05:04:51.978919Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:04:51.978919Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pruning Adversarially Robust Neural Networks without Adversarial Examples","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CR","cs.CV"],"primary_cat":"cs.LG","authors_text":"Jennifer Dy, Stratis Ioannidis, Tong Jian, Yanzhi Wang, Zifeng Wang","submitted_at":"2022-10-09T17:48:50Z","abstract_excerpt":"Adversarial pruning compresses models while preserving robustness. Current methods require access to adversarial examples during pruning. This significantly hampers training efficiency. Moreover, as new adversarial attacks and training methods develop at a rapid rate, adversarial pruning methods need to be modified accordingly to keep up. In this work, we propose a novel framework to prune a previously trained robust neural network while maintaining adversarial robustness, without further generating adversarial examples. We leverage concurrent self-distillation and pruning to preserve knowledg"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.04311","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2210.04311/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2210.04311","created_at":"2026-07-05T05:04:51.978979+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.04311v1","created_at":"2026-07-05T05:04:51.978979+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.04311","created_at":"2026-07-05T05:04:51.978979+00:00"},{"alias_kind":"pith_short_12","alias_value":"5UQAZA3UREC5","created_at":"2026-07-05T05:04:51.978979+00:00"},{"alias_kind":"pith_short_16","alias_value":"5UQAZA3UREC53YR5","created_at":"2026-07-05T05:04:51.978979+00:00"},{"alias_kind":"pith_short_8","alias_value":"5UQAZA3U","created_at":"2026-07-05T05:04:51.978979+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.04951","citing_title":"Robustness as Architecture: Designing IQA Models to Withstand Adversarial Perturbations","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5UQAZA3UREC53YR5GNP4UQCAWV","json":"https://pith.science/pith/5UQAZA3UREC53YR5GNP4UQCAWV.json","graph_json":"https://pith.science/api/pith-number/5UQAZA3UREC53YR5GNP4UQCAWV/graph.json","events_json":"https://pith.science/api/pith-number/5UQAZA3UREC53YR5GNP4UQCAWV/events.json","paper":"https://pith.science/paper/5UQAZA3U"},"agent_actions":{"view_html":"https://pith.science/pith/5UQAZA3UREC53YR5GNP4UQCAWV","download_json":"https://pith.science/pith/5UQAZA3UREC53YR5GNP4UQCAWV.json","view_paper":"https://pith.science/paper/5UQAZA3U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.04311&json=true","fetch_graph":"https://pith.science/api/pith-number/5UQAZA3UREC53YR5GNP4UQCAWV/graph.json","fetch_events":"https://pith.science/api/pith-number/5UQAZA3UREC53YR5GNP4UQCAWV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5UQAZA3UREC53YR5GNP4UQCAWV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5UQAZA3UREC53YR5GNP4UQCAWV/action/storage_attestation","attest_author":"https://pith.science/pith/5UQAZA3UREC53YR5GNP4UQCAWV/action/author_attestation","sign_citation":"https://pith.science/pith/5UQAZA3UREC53YR5GNP4UQCAWV/action/citation_signature","submit_replication":"https://pith.science/pith/5UQAZA3UREC53YR5GNP4UQCAWV/action/replication_record"}},"created_at":"2026-07-05T05:04:51.978979+00:00","updated_at":"2026-07-05T05:04:51.978979+00:00"}